The Genome Recombination/Regulation Section is currently focused on the study of mitotic recombination with an emphasis on events initiated by DSBs. Our studies demonstrated that DSB repair is accompanied by a higher DNA synthesis error rate than found in the normal process that duplicates the genome. The analysis of the processes that determine the fidelity of DSB repair is one of the three major continuing topics. We recognized that some of the tools that we developed to study the fidelity of recombination were also applicable to studying the fidelity of retrotransposition, the second major topic in our laboratory. The third topic grows out of both our growing interest and experience with retrotransposition and a hope that we can make a pragmatic contribution to understanding HIV-1. We demonstrated that the reverse transcriptase (RT) of HIV-1 can substitute for some of the activities of Ty1-RT and that hybrid Ty/HIV-1 elements will multiply in yeast. We continue to use this system to identify drugs active against HIV-1 RT and to better understand HIV-1 RT. I. Fidelity of recombination A. Polymerase Errors Our initial experiments studying the fidelity of mitotic recombination relied on monitoring the reversion of point mutations near the site of a DSB repair event. This assay system was performed in a diploid and enabled us to study the roles of known functions defined in other processes in this process. We demonstrated that the majority of the misincorporation mutations associated recombination are made by the translesion polymerase polz encoded by REV3. Recently, we developed an inverted repeat that allows us to study the fidelity of recombination in haploids. This allowed us to screen for additional genes involved in this process. The inverted repeat substrate monitors the fidelity of copying the CAN1 gene contained within the inverted repeat. Errors are detected as cells resistant to canavanine. We use the galactose regulated HO endonuclease to initiate high levels of DSB recombination. Using these substrates we have demonstrated roles for Exo1p (5) and Rad57p (6) in determining the fidelity of the DNA synthesis associated with recombination. We are continuing to study the roles of known replication and recombination genes in the accuracy of the DNA synthesis associated with recombination. B. NHEJ Repair of DSBs in yeast involves competing pathways of homologous recombination and nonhomologous end joining. The inverted repeat recombination substrate revealed an additional pathway for recombination that combines properties of both systems. In the parental strain, most of the DSBs are repaired by homologous recombination without errors. The majority of the errors involve deletion of the inverted repeat or misincorporation mutations and are dependent on the translesion polymerase polz encoded by REV3, defining a novel role for this polymerase. However, we find that in sae2/com1 mutants as many as 50% of all events are accompanied by a gene amplification process that involves a NHEJ event. Many of those events cause deletion of part of the CAN1 gene. We are continuing to study the roles of genes in controlling the relative contribution of these pathways to the accuracy of DSB repair. II. Fidelity of retrotransposition The accuracy with which the genome of a retrovirus or retrotransposon is copied is the combined consequence of the accuracy of transcription and reverse transcription. We combined the tools used in the recombination studies described above with the genetic marker for retrotransposition developed by the Garfinkel laboratory to develop ways to monitor the fidelity of retrotransposition. Our initial experiments demonstrate that about 2% of the retrotransposition events of a Ty element carrying the TRP1 gene result in errors that create trp1 mutations. These include deletions and point mutations. Similarly, retrotransposition of a Ty element carrying a trp1 point mutation results in a 1000-fold increase in the reversion frequency for either a nonsense or a frameshift allele. It is very difficult to determine the relative contribution of these polymerases to the baseline fidelity of retrotransposition. We are looking for mutations in reverse transcriptase or in components of polII that alter the overall fidelity. A. Reverse transcriptase errors Initially we are focusing on mutants of HIV-1 RT that alter the fidelity of the retrotransposition. There is a substantial literature on the fidelity of HIV-1 RT including the consequences of mutations that confer drug resistance and alterations of amino acids near the active site. In addition to monitoring the fidelity of these characterized mutants in our system, we are creating a new library of HIV-1 RT mutants with lowered fidelity from our screen. B. RNA polymerase errors We are challenging the conventional wisdom that the high mutation rate associated with retroviral replication is all attributable to the accuracy of the reverse transcriptase. The fidelity of transcription is hard to measure. In most cases it is measured indirectly by sequencing cDNAs. There is growing evidence that transcription slippage can occur at high frequency at some locations. We are monitoring the consequences of mutations in components of polII on the fidelity of retrotransposition as a way to define the roles of these subunits on transcription fidelity. III. HIV-1 RT The hybrid Ty/HIV-1 elements (HART) that have HIV-1 RT substituted for the Ty1 RT replicate with about 10% of the efficiency of Ty1. They are dependent on homologous recombination because the HIV 1 RT does not prime on the Ty1 signals and hence does not generate the correct ends required for the Ty1 integrase protein. The HART elements are sensitive to the non-nucleoside class of HIV-1 RT inhibitors (NNRTI), while HART elements containing HIV-1 RT from NNRTI resistant viruses are themselves resistant. We are using these properties to identify new NNRTI drugs and to screen for variants that are resistant to new drugs. As noted above, these elements are also being used to study the features of HIV-1 RT that determine its fidelity.